********************************************
Replication files for "Predictable Unpredictability"

By Nikola L. Datzov  - North Dakota School of Law - nikola.datzov@und.edu
	and 
Jason Rantanen - University of Iowa College of Law - jason-rantanen@uiowa.edu
********************************************



* Codebooks
********************************************
Datzov & Rantanen - Codebook for 101 decisionmaking 2024-11-20.docx 
	Describes variables and how they were coded/created.
	
	
* Analysis Datasets and Code
********************************************
2023-12-31 CAFC documents Release.tab 
	This is the base Federal Circuit decisions dataset dataset for the project. 

Datzov & Rantanen - Predictable Unpredictability 101 Fed. Cir. Cases Dataset - 11.20.24.xlsx 
	This is the dataset of 101 cases with the the human-coded data.

Judge Vote Code.ipynb
	This Jupyter Notebook uses the '101 Fed. Cir. Cases Database' file to generate judge binary vote variables. It outputs a file called "judge votes.csv".

Datzov & Rantanen - 101 Judge Votes 2024-11-20.xlsx 
	This dataset contains binary variables for judge votes generated using Judge Vote Code.ipynb along with the other fields.

Prepare Data.do
	This State code prepares the 101 Judge Votes 2024-11-20.xlsx file for analysis. It outputs a file named ""2024-11-20 101 cases.dta".

Data Analysis.do
	This Stata code uses the "2024-11-20 101 cases.dta" and "2022-12-31 CAFC documents release' datasets to generate reported data and figures for the paper. It can be run immediately after Prepare data.do. Note that the CAFC documents Release must be imported and saved in dta format for it to run.  

* Notes
********************************************
The Stata .do files will require installation of asdoc and tab_chi modules to work properly. To install, run "sec install" followed by the module name. If something isn't running, check to make sure that the files are correctly named and in the correct directory. 




